Re: regression tests

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Jeremy Harris <jgh(at)wizmail(dot)org>
Cc: "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: regression tests
Date: 2013-09-06 18:23:45
Message-ID: CA+TgmoYEOnwDGsi-0b3MBfvkSqUtec+EU7Z5ge8LMb0itaXwbA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, Sep 6, 2013 at 1:34 PM, Jeremy Harris <jgh(at)wizmail(dot)org> wrote:
>> What's an index-hash operation?
>
> Ones that hit tuplesort_begin_index_hash()

Oh. Well, it looks to me like that function can only get called when
building a hash index. Specifically, according to the comment in
hashbuild(), a hash index projected to be larger than shared_buffers.
The regression tests are generally designed to work on small amounts
of data since they need to run quickly, so this isn't too surprising.
Hash indexes are a somewhat underwhelming feature anyway, since btrees
typically perform as well or better, and since hash indexes are not
WAL-logged and therefore can be corrupted on a crash.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Hannu Krosing 2013-09-06 18:34:46 Re: [HACKERS] Re: [HACKERS] Re: [HACKERS] Re: [HACKERS] Is it necessary to rewrite table while increasing the scale of datatype numeric?
Previous Message Robert Haas 2013-09-06 17:57:52 Re: [HACKERS] Re: [HACKERS] Re: [HACKERS] Is it necessary to rewrite table while increasing the scale of datatype numeric?